Event Captioning on YouCook2 1.0 (val)
12.8METEORHiCM²
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| HiCM²Pretraining=Yes, Backbone=CLIP2024.12 | 12.8 | 71.84 | 6.11 | 10.73 | |
| Vid2SeqPretraining=Yes, Backbone=CLIP, Reproduced=true2024.12 | 12.41 | 66.29 | 5.64 | 9.87 | |
| Vid2SeqBackbone=CLIP, Pre-training Videos (# PT)=1M2024.04 | 9.3 | 47.1 | — | 7.9 | |
| DIBSPretraining=Yes, Backbone=CLIP2024.12 | 7.51 | 44.44 | — | 6.39 | |
| Streaming V2SPretraining=Yes, Backbone=CLIP2024.12 | 7.1 | 32.9 | — | 6 | |
| CM2Backbone=CLIP2024.04 | 6.08 | 31.66 | 1.63 | 5.34 | |
| CM²Pretraining=No, Backbone=CLIP2024.12 | 6.08 | 31.66 | 1.63 | 5.34 | |
| PDVCBackbone=CLIP, Reproduced=true2024.04 | 5.56 | 29.69 | 1.4 | 4.92 | |
| PDVCPretraining=No, Backbone=CLIP2024.12 | 5.56 | 29.69 | 1.4 | 4.92 | |
| MTBackbone=TSN2024.04 | 5 | 9.3 | 1.15 | — | |
| ECHRBackbone=C3D2024.04 | 3.82 | — | — | — | |
| E2ESGBackbone=C3D2024.04 | 3.5 | 25 | — | — |